Juvenile myelomonocytic leukemia (JMML) is a disease that occurs in young children and is associated with a high mortality rate. In most patients, JMML has a progressive course leading to death by virtue of infection, bleeding or progression to acute myeloid leukemia (AML). As it is known that children with Neurofibromatosis type 1 (NF1) have a markedly increased risk of developing JMML, we were able to develop a mouse model of JMML by reconstituting lethally irradiated mice with hematopoetic stem cells homozygous for a loss of function mutation in the Nfl gene. In the course of these experiments, we found that all these genetically identical reconstituted mice developed a JMML-like disorder, but only a subset went on to develop more acute disease. This result strongly suggests that additional genetic lesions are responsible for disease progression. The focus of this proposal is to identify these additional genetic lesions as a means to better understand leukemic progression. Toward this goal, we have placed the Nf1 mutation on the BXH-2 mouse genetic background, a strain known to contain a somatically infectious ecotropic retrovirus. Using this powerful somatic mutagenesis system, we have identified three common ecotropic proviral integration (Epi) sites. We hypothesize that these Epi sites will allow identification of genes that are involved in myeloid tumor progression. We have four specific aims: Specific Aim 1:Determine if viral integration at the Epil site leads to deregulation of the c-myb gene. Specific Aim 2:Characterize the gene interrupted by viral integrations at the Epi2 locus. Specific Aim 3: Characterize the gene interrupted by viral integrations at the Epi3 locus. Specific Aim 4: Identify additional Epi sites involved in tumor progression of JMML.